On the Briançon-Skoda theorem on a singular variety (Q978562)

The well known theorem of Briançon-Skoda in his original form [\textit{H. Skoda} and \textit{J. Briançon}, C. R. Acad. Sci., Paris, Sér. A 278, 949--951 (1974; Zbl 0307.32007)] concerns the so-called reduction of an ideal \({\mathfrak A}\) in the local ring \({\mathcal O}_o\) of the holomorphic functions at the origin of \({\mathbb C}^d\). Being an example of a pure algebraic theorem proved by transcendental methods, it attracted the interest of the specialists in algebra \textit{C. Huneke} [Invent. Math. 107, No. 1, 203--223 (1992; Zbl 0756.13001)], \textit{E. Hyry} and \textit{U. O. Villamayor} [J. Algebra 204, 656--665 (1998; Zbl 0928.13006)], \textit{J. Lipman} and \textit{A. Sathaye} [ibid. 28, 199--222 (1981; Zbl 0438.13019)], \textit{J. Lipman} and \textit{B. Teissier} [ibid. 28, 97--112 (1981; Zbl 0464.13005)], \textit{M. Andersson} [Mich. Math. J. 54, No. 2, 361--373 (2006; Zbl 1155.32004)]. More precisely, let \(Z\) be a germ of an analytic space of pure dimension and \({\mathcal O}_2\) be its structure ring of holomorphic functions. Huneke proved that there is a natural number \(\mu\) depending only of \(Z\) such that for any ideal \({\mathfrak A} \subset {\mathcal O}_2\), \({\mathfrak A} = \{ a \}\), and all integers \(l\), \(l\geq 1\), the inclusion \({\mathfrak A}^{\mu + l -1} \subset {\mathfrak A}^l\) holds. In the paper under the review the authors provide a pure analytic proof in which the number \(\mu\) is related with a free resolution of \({\mathcal O}_2\), satisfying the condition that (i) For any ideal \({\mathfrak A} = (a)\), \(a = (a_1, a_2, \ldots , a_n)\), in \({\mathcal O}_Z\) and \(\Phi \subset {\mathcal O}_Z\) the inequality \(| \varphi | \leq C | {\mathfrak A}|^{\mu+l-1}\) implies \(\Phi \in {\mathfrak A}^l\) and (ii) if for a given \({\mathfrak A} = (a_1, a_2, \ldots a_n)\) \(\text{ codim } (Z^r \cap Z^{\mathfrak A}) \geq m+1+r\), \(r \geq 0\) then for any \(\Phi \in {\mathcal O}_2\), \(| \Phi | \leq C | a |^{m+l-1}\) implies \(\Phi \in {\mathcal A}^l\). Clearly, the first part (i) implies the Huneke Theorem. The proof is very technical using different technical tools like modification of the Koszul complex etc.